By Prof. Dr. Leslie Charles Lai Chin Loy Clinical Consultant (Clinical Biochemistry)
Whilst thyroid stimulating hormone (TSH), free triiodothyronine (fT3) and free thyroxine (fT4) are commonly requested biochemical investigations not many doctors are aware of the possible interferences which may occur in the measurement of TSH, fT3 and fT4. These interferences may lead to errors of diagnosis and management. Below are three case reports which illustrate the errors of diagnosis and / or management due to interferences in the measurement of TSH, fT3 and fT4.
A 70 year old lady who has rheumatoid arthritis and primary hypothyroidism had been on thyroxine 100 ug daily for many years. Her TSH had been maintained between 1 and 2 mlU/L all this time. In November 2004, her TSH increased to 8.6 mlU/L but her fT3 and fT4 were normal. Her thyroxine was increased by the Consultant Rheumatologist to 150 ug alternating with 100 ug daily. Two months later her TSH was 18 mlU/L but her fT3 and fT4 were normal. She had also lost 3 kg in weight in the two months.
She was then referred to a Consultant Endocrinologist for management of her hypothyroidism. Although her TSH had risen she had lost weight. The Consultant Endocrinologist held the opinion that she was clinically hyperthyroid although her TSH was high and decreased her thyroxine to 100 ug daily. When she was seen a month later she had stopped losing weight, felt well and was clinically euthyroid. However, her thyroid profile was even more confusing:
fT3 1.2 pmol/L (2.2 - 5.4 pmol/L)
fT4 5.8 pmol/L (9.0-24.0 pmol/L)
TSH 67 mlU/L (0.4 - 4.5 mlU/L)
The biochemical picture indicated that the patient was severely hypothyroid and was not receiving sufficient thyroxine replacement which was not supported by the clinical findings. Prolactin is normally raised in primary hypothyroidism due to the high thyrotrophin-releasing hormone (TRH) level stimulating prolactin seretion. This patient's serum prolactin was thus measured to lend support to the clinical impression that she was on adequate thyroxine replacement therapy. Her serum prolactin was reassuringly normal.
A 35 year old man developed hyperthyroidism in 1999. He was treated with carbimazole for 18 months and the treatment was then stopped. He had then remained clinically and biochemically euthyroid until April 2004 when he was noted to have a raised TSH. He was referred to a Consultant Endocrinologist who made a diagnosis of Hashimoto's disease and commenced him on thyroxine. His TSH, however, remained above 75 mlU/L. His thyroxine was increased to 300 ug daily but his TSH remained above 75 mlU/L while his fT3 and fT4 were normal. He felt very unwell whenever he took the thyroxine so he did not take it regularly. He sought a second opinion in February 2005. He was clinically euthyroid. His fT3 and fT4 were normal but his TSH was 75 mlU/L. A 1 in 5 dilution of his serum sample was made. The TSH after correcting for the dilution was 76 mlU/L. His serum prolactin was normal. His TSH receptor antibodies were > 40.0 IU/L (reference range < 1.0 IU/L). The titres of antimicrosomal and antithyroglobulin antibodies were 1600 (reference range < 100) and < 100 (reference range < 100) respectively. It is very unlikely that he has Hashimoto's disease. His hyperthyroidism in 1999 was due to Graves' disease. His thyroxine was stopped in February 2005 and he has remained well since. He has not had a relapse of his Graves' disease although his TSH receptor antibodies are markedly elevated.
A 32 year old lady developed hyperthyroidism in 2001. She was treated with propylthiouracil for 18 months and her treatment was stopped. She consulted a Consultant Endocrinologist in March 2004 complaining of loss of weight, palpitations and being hot and sweaty. On examination, her blood pressure was 140/90, pulse rate was 112/min. She exhibited increased fine tremor of her hands and she had a large goitre with a thyroid bruit. Apart from lid retraction and lid lag she had no other eye signs. Her hyperthyroidism had relapsed but her thyroid profile suggested that she had primary hypothyroidism:
fT3 1.3 pmol/L (2.2 - 5.4 pmol/L)
fT4 4.2 pmol/L (9.0 - 24.0 pmol/L)
TSH 23 mIU/L (0.4 - 4.5 mIU/L)
She was commenced on propylthiouracil 100 mg three times a day and she felt better. The dosage of propylthiouracil was decreased based on her clinical response. Her TFTs could not be used to monitor her response to treatment as her fT3 and fT4 remained low and her TSH remained elevated.
The high TSH levels in all three cases are due to the presence of heterophilic antibodies to mouse antibody (HAMA) in the sera of these patients. The TSH assay uses a mouse monoclonal antibody. HAMA interferes positively with the measurement of TSH resulting in elevated TSH results. HAMA may develop when the patient comes into contact with mice or even hamsters. The low fT3 and fT4 levels in cases 1 and 3 are due to antibodies to T3 and T4 resulting in a negative interference in the measurement of fT3 and fT4.
The following are ways to test for possible interferences in the measurement of TSH:
- Repeat the measurement using a different immunoassay that uses antibody raised to a different species e.g. sheep.
- Measurement of dilutions of the sample using the manufacturer's diluent containing non-immune globulin.
- Sample pre-treatment e.g. polyethylene glycol precipitation, addition of blocking agents from the same species as the antibody reagents (non-immune serum).
The method to confirm that antibodies to T3 and T4 are present in the patient's serum is to incubate the patient's serum with radioiodine-labelled T3 and T4 respectively and to then precipitate these radiolabelled T3 or T4-antibody complexes by centrifugation.
Immunoassay interference in the measurement of TSH and thyroid hormones may lead to errors of diagnosis and management. If the TSH and thyroid hormone levels are not compatible with the clinical symptoms and signs of the patient the clinician is advised to discuss the case with the biochemist in the laboratory.